Large-scale video generation model developed in China now accessible worldwide.

ShengShu Technology and Tsinghua University developed Vidu, a big video generating model with text-to-video and image-to-video capabilities, which is now available globally.


Vidu can create 4-second clips in 30 seconds and videos up to 32 seconds long in a single instance.

“Vidu can imitate the true physical environment by creating detailed scenarios that follow physical principles such as natural lighting, shadow effects, and intricate face expressions. Furthermore, it can produce surrealistic content with depth and complexity,” stated Zhu Jun, deputy head of the Tsinghua Institute for Artificial Intelligence.

Zhu went on to say that Vidu can build scenarios that capture the essence of many genres such as sci-fi, romance, and animation, as well as high-quality cinematic effects like smoke and lens flares.

The AI model can handle a wide range of shot types, including long shots, close-ups, and medium shots, as well as create effects such as lengthy takes, focus pulls, and smooth scene transitions.

Users can upload photographs or customized character images and utilize written descriptions to direct the characters’ actions in any scene. This feature simplifies the video production process while increasing creative freedom.

The business stated that Vidu’s basic design was proposed as early as 2022.